Menu

Explorer & Settings

Tempo Explorer Submit Project
Back to all projects
A

Antigravity-Super-Skill-Architecture

by scaleto · Updated Jul 5, 2026

A Federated Multi-Agent System (MAS) architecture for Google Antigravity. Orchestrates specialized AI Skills (Backend, DevOps, Frontend) using the Google ADK and A2A Protocol.

5
Stars
1
Forks
Python
Language
Jan 25, 2026
Created

In the AI payments ecosystem

Antigravity-Super-Skill-Architecture is an early-stage Python project in the AI payments / x402 ecosystem, focused on a2a, adk, agent-to-agent, ai-agents. It currently has 5 GitHub stars and 1 forks, and sits alongside related tools like tap, MaiHH-Connect, squidbay.

README.md View on GitHub →

Agent Skill Conglomerate (Super-Skill)

An ecosystem of interconnected "Expert Agents" using a federated cellular architecture, designed to power AI-assisted software development.

🚀 What is this?

This repository contains the implementation of a Super-Skill Architecture, where hundreds of specialized skills (Skills) are organized into 9 Cells (Groups), coordinated by a Master Architect.

Key Features

  • Cellular Architecture: Skills grouped by domain (Backend, Frontend, DevOps, etc.).
  • A2A Protocol: Agent-to-Agent communication system for inter-group collaboration.
  • Intelligent Orchestration: A master-architect that decomposes complex tasks and delegates to experts.

🧠 Global Memory System (Project Context)

Instead of isolated actions, we implemented a central "Source of Truth" where all groups read and write shared information.

  • Why: Ensures the Frontend Group knows instantly which stack the Backend chose without asking, or recalls business decisions made weeks ago.
  • Usage: Automatically managed by the project-memory skill. All major decisions are logged in global_context.yaml.

📄 Standardized Global Templates

A library of professional formats located in 99-common-utils/templates.

  • Why: Guarantees quality and consistency. Whether you ask for an API design or a UI component, the output always follows the same high standard.
  • Usage: The system automatically pulls the correct template for the task (e.g., using api-design.md for backend tasks).

📊 Automatic Status Dashboard

A built-in capability in the Master Architect to generate a "Project Health" report.

  • Why: Provides a quick, automated view of which specific groups have advanced and what tasks remain pending.
  • Usage: Simply ask the Master Architect: "Generate a status report" to receive an updated PROJECT_STATUS.md.

📂 Structure

skills/
├── 00-master-architect/    # The Central Brain
├── 10-backend-group/       # Specialists in API, DB, Servers
├── 20-frontend-group/      # Specialists in UI, React, Mobile
├── 30-devops-group/        # Cloud, CI/CD, K8s
├── 40-security-group/      # Auditing, Pentesting
├── 50-data-ml-group/       # AI, Data Science, RAG
├── 60-testing-group/       # QA, TDD, Code Review
├── 70-docs-group/          # Technical Documentation
├── 80-marketing-group/     # SEO, Growth
├── 85-business-group/      # Strategy, Startups
└── 99-common-utils/        # Global Utilities (A2A, Templates)

🏁 Getting Started

To learn how to use this system, check out:

🛠️ Installation

  1. Clone this repository.
  2. Configure your AI agent (Antigravity/Gemini) to index the skills/ directory.
  3. Ready! Invoke the architect with: "Act as the Master Architect..."

✨ Extensibility

Current capabilities are just the beginning. You can extend the Super-Skill ecosystem on the fly:

How to add new skills: Simply ask Google Antigravity: *"Create a new Skill for [functionality], add it to the [appropriate] group, and integrate it into the Super-Skill environment."*

The architecture is designed to organically grow as you add more specialized agents.

📜 License

MIT License

™️ Credits

This project is an Orchestration Architecture built upon community work. 99% of the individual Skills that make up the cells were adapted from the following amazing repositories:

Thanks to them for creating the fundamental building blocks!